Functional classes of neurons in primary auditory cortex of the cat distinguished by sensitivity to sound location

Abstract

Could neurons in the primary auditory cortex (AI) encode the spatial location of sounds? Is location sensitivity a parameter of stimulus coding relevant to topographical organization in AI? We have mapped receptive fields of single units in AI of the cat by varying systematically the location of a tonal stimulus within a sound field free of acoustic obstructions and reflections. By orienting tangential electrode penetrations parallel to the mediolateral axes along which neurons are tuned to the same stimulus frequency, we have explored the possibility that this axis could be concerned with spatial features of sound stimuli. Approximately half of the neurons encountered were selective for the location of sounds. The location-selective units could be divided into two discrete populations. Hemifield units responded only to sounds presented in the contralateral sound field, with receptive fields extending from a well defined medial border to beyond the contralateral pole of the sound field. Axial units had small, completely circumscribed receptive fields that were constant in location for all such units in each cortical hemisphere studied. The location of axial receptive fields coincided with acoustical 'axis' of the contralateral pinna which was defined by acoustical measurements of the directionality of the ear. Axial units were restricted to the rostral pole of AI where neurons are tuned for high frequencies, while hemifield units were located more caudally and were tuned for lower frequencies. The location-insensitive, omnidirectional, units were distributed across the entire length of AI sampled. Some of the properties of the measured receptive fields could be inferred by comparing the passive acoustical properties of the ear with binaural interactions previously described in AI. There was no indication of a systematic map of sound space in AI. Instead, the mediolateral axis of AI contained spatially segregated regions of different unit classes. This organization was consistent with the previously described pattern of binaural interaction bands. The location-selective and location-insensitive unit classes might represent stages of fundamentally different lines of processing that are segregated within AI.